You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by coderxiang <gi...@git.apache.org> on 2014/06/10 00:22:10 UTC
[GitHub] spark pull request: SPARK-2085: [MLlikb] Apply user-specific regul...
GitHub user coderxiang opened a pull request:
https://github.com/apache/spark/pull/1026
SPARK-2085: [MLlikb] Apply user-specific regularization instead of uniform regularization in ALS
The current implementation of ALS takes a single regularization parameter and apply it on both of the user factors and the product factors. This kind of regularization can be less effective while users number is significantly larger than the number of products (and vice versa). For example, if we have 10M users and 1K product, regularization on user factors will dominate. Following the discussion in [this thread](http://apache-spark-user-list.1001560.n3.nabble.com/possible-bug-in-Spark-s-ALS-implementation-tt2567.html#a2704), the implementation in this PR will regularize each factor vector by #ratings * lambda.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/coderxiang/spark als-reg
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/spark/pull/1026.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1026
----
commit c28d1f9a7ec8b812300b83187199ef7043a5b15c
Author: Shuo Xiang <sx...@twitter.com>
Date: 2014-06-09T22:13:25Z
Apply user-specific regularization instead of uniform regularization in Alternating Least Squares (ALS)
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45577911
All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/15604/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45961793
All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/15735/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45959456
Merged build started.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45958579
All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/15729/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45962156
All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/15736/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/1026#discussion_r13579821
--- Diff: mllib/src/main/scala/org/apache/spark/mllib/recommendation/ALS.scala ---
@@ -533,7 +538,7 @@ class ALS private (
// Add regularization
var i = 0
while (i < rank) {
- fullXtX.data(i * rank + i) += lambda
+ fullXtX.data(i * rank + i) += numRatings(index) * lambda
--- End diff --
minor: you can move `numRatings(index) * lambda` outside the while loop.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45575615
Jenkins, test this please.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45961575
Merged build finished. All automated tests passed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45954951
@coderxiang The merge was not clean. It contains changes from my PR. Could you re-merge the latest master and check the diff is correct on this page?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:
https://github.com/apache/spark/pull/1026
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45577910
Merged build finished. All automated tests passed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45575870
Merged build triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45955129
Merged build started.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45835239
@coderxiang Could you merge the current master? This is not mergeable due to a recent commit.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45955113
Merged build triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45957221
Merged build triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45954727
Jenkins, add to whitelist.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45959101
Merged build started.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45964506
LGTM. Thanks!
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45957669
Merged build started.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45959447
Merged build triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45962155
Merged build finished. All automated tests passed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45551999
Can one of the admins verify this patch?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45961792
Merged build finished. All automated tests passed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45961579
All automated tests passed.
Refer to this link for build results: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/15732/
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/1026#discussion_r13578603
--- Diff: spark ---
@@ -0,0 +1 @@
+Subproject commit 91a64eba90032729873675c6f9c0e53b76b5cf56
--- End diff --
Remove this line.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45575878
Merged build started.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by srowen <gi...@git.apache.org>.
Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45558366
FWIW I support this change. It matches how I have used ALS and is the "weighted regression" (WR) in "ALS-WR" from the paper http://www.hpl.hp.com/personal/Robert_Schreiber/papers/2008%20AAIM%20Netflix/netflix_aaim08(submitted).pdf
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by mengxr <gi...@git.apache.org>.
Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45954737
Jenkins, test this please.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45958578
Merged build finished. All automated tests passed.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---
[GitHub] spark pull request: SPARK-2085: [MLlib] Apply user-specific regula...
Posted by AmplabJenkins <gi...@git.apache.org>.
Github user AmplabJenkins commented on the pull request:
https://github.com/apache/spark/pull/1026#issuecomment-45959085
Merged build triggered.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---